The United States has suffered from a record number of billion dollar natural disasters this year, from the Mississippi and Missouri River flooding to Hurricane Irene and the Texas drought. With each of these events, many have wondered - did climate change have anything to do with this?
Reporters and members of the public wanted to know: has climate change vaulted us into a new normal, as some climate scientists have warned for years, in which these kinds of devastating events will occur more frequently, and be more severe? Or is this just one awful and anomalous year?
Depending on the extreme event, climate science has provided some answers, but they have contained significant caveats, largely because the links between climate change and specific extreme weather events are tenuous, and only now being investigated using methods that are showing greater promise.
Consider, for a moment, where climate science could be in the not too distant future, though.
Imagine a future in which every major extreme weather event is followed up in near real-time by a rigorous scientific assessment of if, and how, manmade global warming influenced the odds that such an event would take place. Such assessments would have numerous benefits, since they would quantify how climate change is shifting the odds of damaging extreme weather events, such as those we’ve seen this year.
They assessments would be prized by basically anyone who needs to make decisions based on the probabilities that a particular threat, such as a 500-year flood, will occur. That’s a lot of people. These studies could also make for excellent opportunities to communicate climate science to the public.
As reported by Quirin Schiermeier in the journal Nature , an international group of climate scientists is working to coordinate and advance studies of climate change attribution, leading us closer to such a real-time attribution system. A white paper on their effort, known for the time being as “ACE”, which stands for “attribution of climate-related events,” is expected to be presented to a climate conference in Denver, Colorado next month.
There have already been numerous studies investigating links between trends in extreme weather events and global climate change. The ACE effort seeks to build on methods used in two key studies published this year, using computer modeling and statistical methods known as “fractional attribution.” One showed that manmade climate change very likely increased the odds of a major flood in the U.K. in 2000, and the other found that manmade climate change likely is responsible for some of the late 20th century increase in Northern Hemisphere intense rainfall events.
However, the science of climate change attribution is still in its infancy, as noted in the Nature story and an accompanying editorial. Methods have advanced enough so that attribution studies are a viable area of focus within climate science, one that holds particular promise for communicating climate science to the public.
This year, at least, the question of “What role did climate change play in (insert a specific extreme event here)?” seems more relevant than ever.
As the Nature editorial states:
The question, after all, seems fair, given the dire warnings of worsening weather that are offered to the public as reasons to care about global warming… Most people associate the climate with the weather that they experience, even if they aren’t supposed to. And they are right to wonder how and why that experience can, on occasion, leave their homes in pieces…
Climate scientists, too, have an obligation to provide more coherent answers to queries (or doubts) as to how global warming influences our weather. An attribution system with ample resources, running in near real time, could prevent scientists’ answers to those questions seeming either too cautious or too alarmist and speculative. It could also prevent the public from getting the (false) impression that climate research is confined to the virtual world of climate models and has little to offer when it comes to current reality, or that climate science is a quasi-experimental field that yields scary but mostly unverifiable results.
Of course, there are major caveats involved with attribution studies, largely centering on the reliability of long-term observational data and computer modeling capabilities. As the Nature editorial states (and as is discussed in more detail in Schiermeier’s story):
Attribution is only as good as the models and statistics that power it - and the various existing climate models project different trends in future extremes in some regions. There is a lack, or poor availability, of long-term observational records, and of climate data with high spatial and temporal resolution. And however it develops, climate attribution will remain rooted in probabilities. Not even the most thorough study can work out with absolute confidence the exact fingerprint of global warming in a given weather event.
Such uncertainties are enough to make some climate scientists queasy. Judith Curry, an outspoken critic of the U.N. Intergovernmental Panel on Climate Change (IPCC) and a force within the climate blogosphere, attacked the ACE effort while heaping praise on a group within the National Oceanic and Atmospheric Administration (NOAA) that has relied more on historical weather observations to put extreme events into context.
While the ACE researchers, led by Peter Stott of the UK Met Office, see their work as responding to a legitimate - and growing - public need for information, Curry sees political motivations as the primary driver of such work.
“Never let a good disaster go to waste in terms of trying to play on people’s emotions to generate support for CO2 mitigation policies,” she wrote on her blog.
Curry is also harshly critical of using computer models in attribution studies.
The NOAA group examines the historical data record, looks for past analogues, and interprets the extreme weather event in the context of the weather and climate dynamics. In the three examples of “What’s happening now”, they explained the Russian heatwave, the snowy winter, and the tornado outbreak in the context of natural weather and climate variability. By contrast, the ACE effort is focused on using climate models to assess the fraction of the event that might be attributed to global warming. It is the use of climate models for this exercise that I object to...
Stott and other attribution researchers readily admit that improvements are needed in computer modeling, statistical techniques, as well as weather and climate observations, in order to make the attribution studies more accurate.
And that’s more than sufficient cause to pursue these studies. If anything, it’s extra motivation for the scientists and funders of this work to push forward and advance the state of the science.
Related: In-depth piece by my Climate Central colleague, Claudia Tebaldi, on the science of extreme event attribution